Disseminate: The Computer Science Research Podcast

Mateusz Gienieczko | AnyBlox: A Framework for Self-Decoding Datasets | #69

In this episode of Disseminate: The Computer Science Research Podcast, host Dr. Jack Waudby is joined by Mateusz Gienieczko, PhD researcher at TU Munich and co-author of the VLDB Best Paper Award winning paper AnyBlox.They dive deep into a fundamental problem in modern data systems: why cutting-edge data encodings and file formats rarely make it from research into real-world systems — and how AnyBlox proposes a radical solution.Mateusz explains the core idea of self-decoding data, where datasets ship with their own portable, sandboxed decoders, allowing any database system to read any encoding safely and efficiently. Built on WebAssembly, AnyBlox bridges the long-standing gap between database research and practice without sacrificing performance, portability, or security.This episode is essential listening for database researchers, data engineers, system builders, and industry practitioners interested in the future of data formats, analytics performance, and making research matter in practiceLinks:Paper: https://www.vldb.org/pvldb/vol18/p4017-gienieczko.pdfGitHub: https://github.com/AnyBloxMat's Homepage: https://v0ldek.com/ Hosted on Acast. See acast.com/privacy for more information.

Xiangyao Yu | Disaggregation: A New Architecture for Cloud Databases | #68

In this episode of Disseminate: The Computer Science Research Podcast, host Jack Waudby sits down with Xiangyao Yu (UW–Madison), one of the leading voices shaping the next generation of cloud-native databases.We dive deep into disaggregation — the architectural shift transforming how modern data systems are built. Xiangyao breaks down:Why traditional shared-nothing databases struggle in cloud environmentsHow separating compute and storage unlocks elasticity, scalability, and cost efficiencyThe evolution of disaggregated systems, from Aurora and Snowflake through to advanced pushdown processing and new modular servicesHis team's research on reinventing core protocols like 2-phase commit for cloud-native environmentsReal-time analytics, HTAP challenges, and the Hermes architectureWhere disaggregation goes next — indexing, query optimizers, materialized views, multi-cloud architectures, and moreWhether you're a database engineer, researcher, or a practitioner building scalable cloud systems, this episode gives a clear, accessible look into the architecture that’s rapidly becoming the default for modern data platforms.Links:Xiangyao Yu's HomepageDisaggregation: A New Architecture for Cloud Databases [VLDB'25] Hosted on Acast. See acast.com/privacy for more information.

Navid Eslami | Diva: Dynamic Range Filter for Var-Length Keys and Queries | #67

In this episode of Disseminate: The Computer Science Research Podcast, Jack sits down with Navid Eslami, PhD researcher at the University of Toronto, to discuss his award-winning paper “DIVA: Dynamic Range Filter for Variable Length Keys and Queries”, which earned Best Research Paper at VLDB.Navid breaks down how range filters extend the power of traditional filters for modern databases and storage systems, enabling faster queries, better scalability, and theoretical guarantees. We dive into:How DIVA overcomes the limitations of existing range filtersWhat makes it the “holy grail” of filtering for dynamic dataReal-world integration in WiredTiger (the MongoDB storage engine)Future challenges in data distribution smoothing and hybrid filteringWhether you're a database engineer, systems researcher, or student exploring data structures, this episode reveals how cutting-edge research can transform how we query, filter, and scale modern data systems.Links:Diva: Dynamic Range Filter for Var-Length Keys and Queries [VLDB'25]Diva on GitHubNavid's LinkedIn Hosted on Acast. See acast.com/privacy for more information.

Adaptive Factorization in DuckDB with Paul Groß

In this episode of the DuckDB in Research series, host Jack Waudby sits down with Paul Groß, PhD student at CWI Amsterdam, to explore his work on adaptive factorization and worst-case optimal joins - techniques that push the boundaries of analytical query performance.Paul shares insights from his CIDR'25 paper “Adaptive Factorization Using Linear Chained Hash Tables”, revealing how decades of database theory meet modern, practical system design in DuckDB. From hash table internals to adaptive query planning, this episode uncovers how research innovations are becoming part of real-world systems.Whether you’re a database researcher, engineer, or curious student, you’ll come away with a deeper understanding of query optimization and the realities of systems engineering.Links:Adaptive Factorization Using Linear-Chained Hash Tables Hosted on Acast. See acast.com/privacy for more information.

Parachute: Rethinking Query Execution and Bidirectional Information Flow in DuckDB - with Mihail Stoian

In this episode of the DuckDB in Research series, host Jack Waudby sits down with Mihail Stoian, PhD student at the Data Systems Lab, University of Technology Nuremberg, to unpack the cutting-edge ideas behind Parachute, a new approach to robust query processing and bidirectional information passing in modern analytical databases.We explore how Parachute bridges theory and practice, combining concepts from instance-optimal algorithms and semi-join filtering to boost performance in DuckDB, the in-process analytical SQL engine that’s reshaping how research meets real-world data systems.Mihail discusses:How Parachute extends semi-join filtering for two-way information flowThe challenges of implementing research ideas inside DuckDBPractical performance gains on TPC-H and CEB workloadsThe future of adaptive query processing and research-driven system designWhether you're a database researcher, systems engineer, or curious practitioner, this deep-dive reveals how academic innovation continues to shape modern data infrastructure.Links:Parachute: Single-Pass Bi-Directional Information Passing VLDB 2025 PaperMihail's homepageParachute's Github repo Hosted on Acast. See acast.com/privacy for more information.